Study evaluates response to psychotherapeutical interventions in anorexia nervosa

This study demonstrates that the distinction between threshold and subthreshold anorexia nervosa does not seem to be of clinical relevance in terms of response to cognitive-behavioral therapy (CBT). Shape concern rather than demographic or general psychopathological features represents the best predictor of outcome for CBT.

Few long-term follow-up studies have evaluated the response to psychotherapeutical interventions in anorexia nervosa (AN). In this study, the effectiveness of individual cognitive-behavioral therapy (CBT) and the possible predictors of outcome in outpatients suffering from threshold and subthreshold AN (s-AN) were evaluated. At the beginning (T0) and at the end of treatment (T1), and 3 years after the end of treatment (T2), 53 subjects with AN and 50 with s-AN (all DSM-IV criteria except amenorrhea or underweight) were assessed by a face-to-face clinical interview and by self-reported questionnaires for eating attitudes and behavior (Eating Disorder Examination Questionnaire), body uneasiness (Body Uneasiness Test) and general psychopathology (Symptom Checklist, Beck Depression Inventory, State-Trait Anxiety Inventory).

No deaths occurred during the treatment and the follow-up period. At the end of the follow-up 34 subjects (33%) initially enrolled in the study obtained a full recovery. AN and s-AN patients did not show significant differences on most of the clinical measures at baseline and in terms of treatment response (T1, T2). The reduction in weight and shape concerns was associated with weight gain at T1 and T2, and the shape concern level at baseline represented the main risk factor for recovery and treatment resistance. According to survival analysis, patients with high shape concern had a lower probability of remission across time.

The distinction between threshold and subthreshold AN does not seem to be of clinical relevance in terms of response to CBT. Shape concern rather than demographic or general psychopathological features represents the best predictor of outcome for CBT.